06 August 2026
A new study, 'Generative design of bacteriophages with genome language models’ by Samuel H. King et al, was published in Science at 1900 UK time today (6 August 2026).
Commenting on the study, Dr Simon Clarke, Associate Professor in Cellular Microbiology, University of Reading, said:
"In this very important publication, backed by excellent quality data, researchers at Stanford University used AI models to design novel variants of phiX174, an E. coli-infecting virus called a bacteriophage. The models generated hundreds of synthetic genomes distinct from any sequences found in nature.
"After synthesizing this genetic material in the lab, experiments confirmed that 16 of the AI-designed phages successfully infected E. coli, exhibiting varied fitness levels; three variants even outcompeted the original parent strain. Crucially, several synthetic viruses overcame natural bacterial resistance to phiX174, supporting the authors' proposal that AI-designed viruses could advance phage therapy to treat antibiotic-resistant infections, which has been the focus of decades of research, with only limited success.
"While the authors previously showed that dataset exclusions can prevent AI models from generating human pathogens, this study proves AI can create viable, non-natural viruses. So, while there is no work to date using this method to generate human pathogens, this capability does raise significant ethical questions as it confirms that the technology to generate fitter viruses is here and can be done more efficiently than before.
"As the authors highlight, this raises some serious regulatory and safety concerns, to say the very least. While work of this nature is normally tightly regulated, it is reassuring that these scientists have shown further restraint in providing important guard rails, but there is no guarantee that every other scientist attempting to do something similar will be so careful.”
Dr Clarke is available for further comment - contact the University of Reading press office on 0118 378 5757 or pressoffice@reading.ac.uk.
We are in the process of finalising our postgraduate taught courses for 2027/28 entry. In the meantime, you can view our 2026/27 courses.
